Unstructured Key Definitions for Optimal Performance

1. A method comprising: running a comment examining process for processing comments of one or more user to provide a comment processing output; applying data of a dataset as training data for training a neural network to define a trained neural network, wherein the training data includes input node training data and output node training data; and generating a decision rule for the dataset based on a transfer function of the trained neural network, wherein the generating includes querying the trained neural network using sample data, wherein the querying is in dependence on one or more topic of interest extracted by the comment examining process.

2. The method of claim 1 , wherein the comment processing output includes user profile information of a certain user, and wherein the method includes presenting the decision rule to the certain user.

3. The method of claim 1 , wherein the comment processing output is a crowdsourced output, and wherein the method includes configuring the neural network based on the crowdsourced output.

4. The method of claim 1 , wherein the comment processing output is a crowdsourced output, wherein the method includes configuring the neural network based on the crowdsourced output, wherein the crowdsourced output includes a prioritized list of comment topics determined using aggregated data of a plurality of users, and wherein the method includes selecting input node parameters for input nodes of the neural network based on the prioritized list.

5. The method of claim 1 , wherein the comment processing output includes user profile information that indicates one or more topic of interest to a certain user, wherein the method includes applying a rule generating process for performing the generating based on the one or more topic of interest to the certain user.

6. The method of claim 1 , wherein the comment processing output includes a crowdsourced output that indicates one or more topic of interest to an aggregate of users, wherein the method includes applying a rule generating process for performing the generating based on the one or more topic of interest to the aggregate of users.

7. The method of claim 1 , wherein the comment processing output includes user profile information that indicates one or more topic of interest to a certain user, wherein the method includes applying a rule generating process for performing the generating based on the one or more topic of interest to the certain user, wherein the rule generating process includes querying the trained neural network using a set of values applied to a certain input node of the neural network, the certain input node being based on the one or more topic of interest.

8. The method of claim 1 , wherein the neural network includes a plurality of residual links, wherein the comment processing output includes user profile information that indicates one or more topic of interest to a certain user, wherein the method includes applying a rule generating process for performing the generating based on the one or more topic of interest to the certain user, wherein the rule generating process includes deactivating links of the plurality of residual links based on the one or more topic of interest.

9. The method of claim 1 , wherein the method includes selecting input node parameters based on a prioritized list of comment topics determined using aggregated data of a plurality of users.

10. The method of claim 1 , wherein the applying data of the dataset as training data for training a neural network to define a trained neural network includes applying the input node training data to an input node of the neural network and applying the output node training data to an output node of the neural network.

11. The method of claim 1 , wherein the comments specify actions of one or more individual engaged in an activity, and wherein the comments are comments of one or more user observing the activity.

12. The method of claim 1 , wherein the sample data is determined in dependence on one or more topic of interest extracted by the comment examining process.

13. The method of claim 1 , wherein the generating the decision rule for the dataset based on the transfer function of the trained neural network includes querying the trained neural network using a set of values to determine a characteristic of the transfer function, wherein the set of values is determined in dependence on one or more topic of interest extracted from the processing comments by the comment examining process.

14. The method of claim 1 wherein the generating the decision rule for the dataset based on the transfer function of the trained neural network includes querying the trained neural network using a spread of values to determine a characteristic of the transfer function, wherein the spread of values is determined in dependence on one or more topic of interest extracted by the comment examining process.

15. The method of claim 1 , wherein the comment processing output includes a crowdsourced output that indicates one or more topic of interest to an aggregate of users, wherein the generating the decision rule for the dataset based on the transfer function of the trained neural network includes querying the trained neural network using a set of values to determine a characteristic of the transfer function, wherein the set of values is determined in dependence on the one or more topic of interest to an aggregate of users extracted from the processing comments by the comment examining process.

16. A method comprising: running a comment examining process for processing comments of one or more user to provide a comment processing output; configuring a neural network based on the comment processing output, wherein training data for training the neural network to define a trained neural network includes input node training data and output node training data; and predicting a result of an event using the trained neural network, wherein the comment processing output is a crowdsourced output, wherein the method includes configuring the neural network based on the crowdsourced output, wherein the crowdsourced output includes a prioritized list of comment topics determined using aggregated data of a plurality of users, and wherein the method includes selecting input node parameters for input nodes of the neural network based on the prioritized list.

17. The method of claim 16 , wherein the comment processing output is a crowdsourced output, and wherein the method includes configuring the neural network based on the crowdsourced output.

18. The method of claim 16 , wherein the comment processing output is a crowdsourced output, wherein the method includes configuring the neural network based on the crowdsourced output, wherein the crowdsourced output includes a prioritized list of comment topics determined using aggregated data of a plurality of users, and wherein the method includes selecting input node parameters for input nodes of the neural network based on the prioritized list.

19. The method of claim 16 , wherein the trained neural network includes a plurality of residual links subject to being deactivated, wherein the predicting includes performing first predicting for sending a first prediction to a first destination and performing second predicting for sending a second prediction to a second destination, wherein performing the first predicting includes deactivating first links of the plurality of residual links, wherein performing the second predicting includes deactivating second links of the plurality of residual links.

20. The method of claim 16 , wherein the comments specify actions of one or more individual engaged in an activity, and wherein the comments are comments of one or more user observing the activity.

21. A method comprising: running a comment examining process for processing comments of one or more user to provide a comment processing output; applying data of a dataset as training data for training a neural network to define a trained neural network, wherein the training data includes input node training data and output node training data; and generating a decision rule for the dataset based on a transfer function of the trained neural network, wherein the decision rule is based on the comment processing output, wherein the comment processing output includes user profile information that indicates one or more topic of interest to a certain user, wherein the method includes applying a rule generating process for performing the generating based on the one or more topic of interest to the certain user, wherein the rule generating process includes querying the trained neural network using a set of values applied to a certain input node of the neural network, the certain input node being based on the one or more topic of interest.